This project is concerned with statistical methodological issues involved in the design and analysis of animal bioassays with particular emphasis on TRTP's two-year carcinogenesis bioassay. Current research efforts are focused on a detailed assessment of the basic experimental design employed in the TRTP cancer bioassay, on the development of statistical procedures for incorporating historical control data in a formal testing framework, and on the evaluation of various factors that may contribute to intra- and inter-laboratory variability in control tumor incidence. In addition the identification and analysis of treatment-associated patterns of change in tumor incidence has also been initiated for recently completed TRTP cancer bioassays.